audio-plot-utils.py

# %% [code]
#-------------------------------------
#
# Audio plot utils:
# - text formater (auxiliar/utils)
# - sound player
# - display sound wave
# - display spectrogram
# - display mel spectrogram
#
#-------------------------------------
importpandasaspd
importnumpyasnp
importmatplotlib.pyplotasplt
importlibrosa
fromscipy.ioimportwavfile
fromlibrosa.displayimportwaveshow
importIPython.displayasipd
fromIPython.displayimportAudio
fromIPython.displayimportdisplay, HTML


defcstr(str_text, color='black'):
"""
 Html styling for widgets
 Args
 str_text: text to disply
 color: color to display the text
 Returns
 Formated text/label
 """
return"<text style=color:{}><strong>{}<strong></text>".format(color, str_text)

defplay_sound(sound_path="",
text="Test", 
color="green"):
"""
 Display a sound play widget
 Args
 sound_path: path to the sound file
 text: text to display
 color: color for text to display
 Returns
 None
 """
display(HTML(cstr(text, color)))
display(ipd.Audio(sound_path))

defdisplay_sound_wave(sound_path=None,
text="Test", 
color="green"):
"""
 Display a sound wave
 Args
 sound_path: path to the sound file
 text: text to display
 color: color for text to display
 Returns
 None
 """
ifnotsound_path:
return
y_sound, sr_sound=librosa.load(sound_path)
audio_sound, _=librosa.effects.trim(y_sound)
fig, ax=plt.subplots(1, figsize= (16, 3))
fig.suptitle(f'Sound Wave: {text}', fontsize=12)
librosa.display.waveshow(y=audio_sound, sr=sr_sound, color=color)

defdisplay_wavefile(sound_path=None,
text="Test", 
color="green"):
"""
 Display a sound wave - load using wavefile
 sr: sample rate
 y_sound: sound samples
 Args
 sound_path: path to the sound file
 text: text to display
 color: color for text to display
 Returns
 None
 """
ifnotsound_path:
return
sr_sound, y_sound=wavfile.load(sound_path)
fig, ax=plt.subplots(1, figsize= (16, 3))
fig.suptitle(f'Sound Wave: {text}', fontsize=12)
ax.plot(np.linspace(0, sr_sound/len(y_sound), sr_sound), y_sound)


defdisplay_spectrogram(sound_path=None,
text="Test"):
"""
 Display a spectrogram
 Args
 sound_path: path to the sound file
 text: text to display (title)
 Returns
 None 
 """
ifnotsound_path:
return
n_fft=2048
hop_length=512
y_sound, sr_sound=librosa.load(sound_path)
audio_sound, _=librosa.effects.trim(y_sound)

# Short-time Fourier transform (STFT)
D_sound=np.abs(librosa.stft(audio_sound, n_fft=n_fft, hop_length=hop_length))
# Convert an amplitude spectrogram to Decibels-scaled spectrogram.
DB_sound=librosa.amplitude_to_db(D_sound, ref=np.max)
# Prepare the plot
fig, ax=plt.subplots(1, 1, figsize=(12, 4))
img=librosa.display.specshow(DB_sound, sr=sr_sound, hop_length=hop_length, x_axis='time', 
y_axis='log', cmap='cool', ax=ax)
ax.set_title(f'Log Frequency Spectrogram: {text}', fontsize=10) 
plt.colorbar(img,ax=ax)

defdisplay_mel_spectrogram(sound_path=None,
text="Test"):
"""
 Display a mel spectrogram
 Args
 sound_path: path to the sound file
 text: text to display (title)
 Returns
 None 
 """
ifnotsound_path:
return
hop_length=512
y_sound, sr_sound=librosa.load(sound_path)
audio_sound, _=librosa.effects.trim(y_sound)
# Create the Mel Spectrograms
S_sound=librosa.feature.melspectrogram(audio_sound, sr=sr_sound)
S_DB_sound=librosa.amplitude_to_db(S_sound, ref=np.max)
# Prepare the plot
fig, ax=plt.subplots(1, 1, figsize=(12, 6))
img=librosa.display.specshow(S_DB_sound, sr=sr_sound, hop_length=hop_length, x_axis='time', 
y_axis='log', cmap='cool', ax=ax)

ax.set_title(f'Mel Spectrogram: {text}', fontsize=10) 
plt.colorbar(img,ax=ax)